Published in:
01-07-2020 | Breast Cancer | Original Article
Breast cancer risk prediction models and subsequent tumor characteristics
Published in: Breast Cancer | Issue 4/2020
Login to get accessAbstract
Background
A previous study found evidence that a breast cancer risk prediction model preferentially selected for less aggressive tumors in Swedish women. In the US, the Gail model has been widely used and was used for entry criteria in two large breast cancer prevention trials. We assessed if higher risk levels from the Gail model were associated with less aggressive tumor characteristics and if risk levels were predictive of mortality and survival.
Methods
We used questionnaire data from women in the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial to calculate Gail risk levels (low < 1.66%; moderate 1.66–2.99%; high ≥ 3.00%). Women aged 55–74 were enrolled between 1993 and 2001 and had detailed information on breast cancer incidence and tumors collected. We calculated breast cancer incidence and mortality rates among all women by risk levels and examined breast cancer survival and tumor characteristics among women diagnosed with breast cancer. We used Chi-squared tests and multivariable logistic regression to assess the association between risk levels and tumor characteristics.
Results
The study population for this analysis included 45,402 women with 1908 cases of breast cancer. Women at high risk were associated with higher risk of breast cancer mortality compared to women with low risk [rate ratio (RR) = 2.29 95% confidence interval (CI) 1.37–3.84)]. Higher risk levels were associated with lobular-type tumors [moderate: adjusted odds ratio (aOR) = 1.57 95% CI 1.13–2.17; high: aOR = 1.78 95% CI 1.25–2.54] but were not associated with any other tumor characteristics or breast cancer survival.
Conclusions
We did not find evidence that higher risk levels from the Gail model are predictive of less aggressive breast cancer tumors.